local-deep-research 0.1.18__tar.gz → 0.1.19__tar.gz

{local_deep_research-0.1.18/src/local_deep_research.egg-info → local_deep_research-0.1.19}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: local-deep-research
-Version: 0.1.18
+Version: 0.1.19
 Summary: AI-powered research assistant with deep, iterative analysis using LLMs and web searches
 Author-email: LearningCircuit <185559241+LearningCircuit@users.noreply.github.com>, HashedViking <6432677+HashedViking@users.noreply.github.com>
 License: MIT License
@@ -71,6 +71,14 @@ Dynamic: license-file
 
 A powerful AI-powered research assistant that performs deep, iterative analysis using multiple LLMs and web searches. The system can be run locally for privacy or configured to use cloud-based LLMs for enhanced capabilities.
 
+<div align="center">
+  <a href="https://www.youtube.com/watch?v=0ISreg9q0p0">
+    <img src="https://img.youtube.com/vi/0ISreg9q0p0/0.jpg" alt="Local Deep Research">
+    <br>
+    <span>▶️ Watch Video</span>
+  </a>
+</div>
+
 ## Quick Start
 
 ```bash
@@ -143,6 +151,8 @@ The package automatically creates and manages configuration files in your user d
 
 ### Default Configuration Files
 
+If you prefere environment variables please refere to this file: https://github.com/LearningCircuit/local-deep-research/blob/main/docs/env_configuration.md
+
 When you first run the tool, it creates these configuration files:
 
 | File | Purpose |
@@ -215,6 +225,10 @@ The `OPENAI_ENDPOINT` provider can access any service with an OpenAI-compatible
 
 The system includes multiple search engines. Some require API keys:
 
+Use .env in config folder if .secrets.toml doesnt work. 
+
+You can also overwrite other settings via environment variables, e.g. to overwrite [web] port setting in settings.toml please use: **LDR_WEB__PORT=8080**
+
 ```toml
 # Add to .secrets.toml
 SERP_API_KEY = "your-serpapi-key-here"        # For Google results via SerpAPI

{local_deep_research-0.1.18 → local_deep_research-0.1.19}/README.md

@@ -2,6 +2,14 @@
 
 A powerful AI-powered research assistant that performs deep, iterative analysis using multiple LLMs and web searches. The system can be run locally for privacy or configured to use cloud-based LLMs for enhanced capabilities.
 
+<div align="center">
+  <a href="https://www.youtube.com/watch?v=0ISreg9q0p0">
+    <img src="https://img.youtube.com/vi/0ISreg9q0p0/0.jpg" alt="Local Deep Research">
+    <br>
+    <span>▶️ Watch Video</span>
+  </a>
+</div>
+
 ## Quick Start
 
 ```bash
@@ -74,6 +82,8 @@ The package automatically creates and manages configuration files in your user d
 
 ### Default Configuration Files
 
+If you prefere environment variables please refere to this file: https://github.com/LearningCircuit/local-deep-research/blob/main/docs/env_configuration.md
+
 When you first run the tool, it creates these configuration files:
 
 | File | Purpose |
@@ -146,6 +156,10 @@ The `OPENAI_ENDPOINT` provider can access any service with an OpenAI-compatible
 
 The system includes multiple search engines. Some require API keys:
 
+Use .env in config folder if .secrets.toml doesnt work. 
+
+You can also overwrite other settings via environment variables, e.g. to overwrite [web] port setting in settings.toml please use: **LDR_WEB__PORT=8080**
+
 ```toml
 # Add to .secrets.toml
 SERP_API_KEY = "your-serpapi-key-here"        # For Google results via SerpAPI

{local_deep_research-0.1.18 → local_deep_research-0.1.19}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "local-deep-research"
-version = "0.1.18"
+version = "0.1.19"
 description = "AI-powered research assistant with deep, iterative analysis using LLMs and web searches"
 readme = "README.md"
 requires-python = ">=3.8"
@@ -66,4 +66,4 @@ include-package-data = true
 
 [tool.setuptools.package-data]
 "local_deep_research.web" = ["templates/*", "static/*", "static/**/*"]
-"local_deep_research.defaults" = ["*.toml", "*.py"]
+"local_deep_research.defaults" = ["*.toml", "*.py", "*.template"]

{local_deep_research-0.1.18 → local_deep_research-0.1.19}/src/local_deep_research/config.py

@@ -145,6 +145,19 @@ def init_config_files():
             if not os.path.exists(search_engines_file) and os.path.exists(default_engines):
                 shutil.copyfile(default_engines, search_engines_file)
                 logger.info(f"Created search_engines.toml at {search_engines_file}")
+
+                    # Create .env.template if it doesn't exist
+            env_template_file = CONFIG_DIR / ".env.template"
+            if not env_template_file.exists():
+                shutil.copy(defaults_dir / ".env.template", env_template_file)
+                logger.info(f"Created .env.template at {env_template_file}")
+                
+                # Optionally create an empty .env file if it doesn't exist
+                env_file = CONFIG_DIR / ".env"
+                if not env_file.exists():
+                    with open(env_file, "w") as f:
+                        f.write("# Add your environment variables here\n")
+                    logger.info(f"Created empty .env file at {env_file}")
         except Exception as e:
             logger.error(f"Error initializing Windows config files: {e}")
     else:
@@ -183,7 +196,17 @@ def init_config_files():
         if not search_engines_file.exists():
             shutil.copy(defaults_dir / "search_engines.toml", search_engines_file)
             logger.info(f"Created search_engines.toml at {search_engines_file}")
+        env_template_file = CONFIG_DIR / ".env.template"
+        if not env_template_file.exists():
+            shutil.copy(defaults_dir / ".env.template", env_template_file)
+            logger.info(f"Created .env.template at {env_template_file}")
             
+            # Optionally create an empty .env file if it doesn't exist
+            env_file = CONFIG_DIR / ".env"
+            if not env_file.exists():
+                with open(env_file, "w") as f:
+                    f.write("# Add your environment variables here\n")
+                logger.info(f"Created empty .env file at {env_file}")            
         secrets_file = CONFIG_DIR / ".secrets.toml"
         if not secrets_file.exists():
             with open(secrets_file, "w") as f:

local_deep_research-0.1.19/src/local_deep_research/defaults/llm_config.py

@@ -0,0 +1,307 @@
+"""
+LLM configuration for Local Deep Research.
+
+This file controls which language models are available and how they're configured.
+You can customize model selection, parameters, and fallbacks here.
+"""
+
+from langchain_anthropic import ChatAnthropic
+from langchain_openai import ChatOpenAI
+from langchain_ollama import ChatOllama
+from langchain_community.llms import VLLM
+from local_deep_research.config import settings
+import os
+import logging
+
+# Initialize environment
+logger = logging.getLogger(__name__)
+
+# Valid provider options
+VALID_PROVIDERS = ["ollama", "openai", "anthropic", "vllm", "openai_endpoint", "lmstudio", "llamacpp", "none"]
+
+# ================================
+# LLM FUNCTIONS
+# ================================
+
+def get_llm(model_name=None, temperature=None, provider=None):
+    """
+    Get LLM instance based on model name and provider.
+    
+    Args:
+        model_name: Name of the model to use (if None, uses settings.llm.model)
+        temperature: Model temperature (if None, uses settings.llm.temperature)
+        provider: Provider to use (if None, uses settings.llm.provider)
+    
+    Returns:
+        A LangChain LLM instance
+    """
+    # Use settings values for parameters if not provided
+    if model_name is None:
+        model_name = settings.llm.model
+    
+    if temperature is None:
+        temperature = settings.llm.temperature
+        
+    if provider is None:
+        provider = settings.llm.provider.lower()
+        if provider not in VALID_PROVIDERS:
+            logger.error(f"Invalid provider in settings: {provider}")
+            raise ValueError(f"Invalid provider: {provider}. Must be one of: {VALID_PROVIDERS}")
+    
+    # Common parameters for all models
+    common_params = {
+        "temperature": temperature,
+        "max_tokens": settings.llm.max_tokens,
+    }
+    
+    # Handle different providers
+    if provider == "anthropic":
+        api_key = settings.get('ANTHROPIC_API_KEY', '')
+        if not api_key:
+            logger.warning("ANTHROPIC_API_KEY not found. Falling back to default model.")
+            return get_fallback_model(temperature)
+        
+        return ChatAnthropic(
+            model=model_name, anthropic_api_key=api_key, **common_params
+        )
+    
+    elif provider == "openai":
+        api_key = settings.get('OPENAI_API_KEY', '')
+        if not api_key:
+            logger.warning("OPENAI_API_KEY not found. Falling back to default model.")
+            return get_fallback_model(temperature)
+        
+        return ChatOpenAI(model=model_name, api_key=api_key, **common_params)
+    
+    elif provider == "openai_endpoint":
+        api_key = settings.get('OPENAI_ENDPOINT_API_KEY', '')
+
+        if not api_key:
+            logger.warning("OPENAI_ENDPOINT_API_KEY not found. Falling back to default model.")
+            return get_fallback_model(temperature)
+        
+        # Get endpoint URL from settings
+        openai_endpoint_url = settings.llm.openai_endpoint_url
+        
+        return ChatOpenAI(
+            model=model_name, 
+            api_key=api_key,
+            openai_api_base=openai_endpoint_url, 
+            **common_params
+        )
+    
+    elif provider == "vllm":
+        try:
+            return VLLM(
+                model=model_name,
+                trust_remote_code=True,
+                max_new_tokens=128,
+                top_k=10,
+                top_p=0.95,
+                temperature=temperature,
+            )
+        except Exception as e:
+            logger.error(f"Error loading VLLM model: {e}")
+            logger.warning("Falling back.")
+            return get_fallback_model(temperature)
+    
+    elif provider == "ollama":
+        try:
+            # Use the configurable Ollama base URL
+            base_url = settings.get('OLLAMA_BASE_URL', settings.llm.get('ollama_base_url', 'http://localhost:11434'))
+            return ChatOllama(model=model_name, base_url=base_url, **common_params)
+        except Exception as e:
+            logger.error(f"Error loading Ollama model: {e}")
+            return get_fallback_model(temperature)
+    
+    elif provider == "lmstudio":
+        try:
+            # Import LM Studio package
+            import lmstudio
+            from langchain_core.language_models import BaseLLM
+            
+            # Get LM Studio URL from settings
+            lmstudio_url = settings.llm.get('lmstudio_url', "http://localhost:1234")
+            
+            # Create LM Studio LLM instance
+            model = lmstudio.llm(model_name)
+            
+            # Return a LangChain compatible wrapper
+            class LMStudioLLM(BaseLLM):
+                def _call(self, prompt, stop=None, **kwargs):
+                    result = model.complete(prompt, temperature=temperature)
+                    return result.completion
+                    
+                @property
+                def _identifying_params(self):
+                    return {"model_name": model_name}
+                    
+                @property
+                def _llm_type(self):
+                    return "lmstudio"
+                    
+            return LMStudioLLM()
+        except ImportError:
+            logger.error("LM Studio package not installed. Run 'pip install lmstudio'")
+            raise ImportError("LM Studio package not installed. Run 'pip install lmstudio'")
+    
+    elif provider == "llamacpp":
+        try:
+            # Import LlamaCpp
+            from langchain_community.llms import LlamaCpp
+            
+            # Get LlamaCpp model path from settings
+            model_path = settings.llm.get('llamacpp_model_path', "")
+            if not model_path:
+                logger.error("llamacpp_model_path not set in settings")
+                raise ValueError("llamacpp_model_path not set in settings.toml")
+                
+            # Get additional LlamaCpp parameters
+            n_gpu_layers = settings.llm.get('llamacpp_n_gpu_layers', 1)
+            n_batch = settings.llm.get('llamacpp_n_batch', 512)
+            f16_kv = settings.llm.get('llamacpp_f16_kv', True)
+            
+            # Create LlamaCpp instance
+            return LlamaCpp(
+                model_path=model_path,
+                temperature=temperature,
+                max_tokens=settings.llm.max_tokens,
+                n_gpu_layers=n_gpu_layers,
+                n_batch=n_batch,
+                f16_kv=f16_kv,
+                verbose=True
+            )
+        except ImportError:
+            logger.error("LlamaCpp package not installed. Run 'pip install llama-cpp-python'")
+            raise ImportError("LlamaCpp package not installed. Run 'pip install llama-cpp-python'")
+    
+    else:
+        return get_fallback_model(temperature)
+
+def get_fallback_model(temperature=None):
+    """Create a dummy model for when no providers are available"""
+    from langchain_community.llms.fake import FakeListLLM
+    return FakeListLLM(
+        responses=["No language models are available. Please install Ollama or set up API keys."]
+    )
+
+# ================================
+# COMPATIBILITY FUNCTIONS
+# ================================
+
+def get_available_provider_types():
+    """Return available model providers"""
+    providers = {}
+    
+    if is_ollama_available():
+        providers["ollama"] = "Ollama (local models)"
+    
+    if is_openai_available():
+        providers["openai"] = "OpenAI API"
+    
+    if is_anthropic_available():
+        providers["anthropic"] = "Anthropic API"
+    
+    if is_openai_endpoint_available():
+        providers["openai_endpoint"] = "OpenAI-compatible Endpoint"
+    
+    if is_lmstudio_available():
+        providers["lmstudio"] = "LM Studio (local models)"
+        
+    if is_llamacpp_available():
+        providers["llamacpp"] = "LlamaCpp (local models)"
+    
+    # Check for VLLM capability
+    try:
+        import torch
+        import transformers
+        providers["vllm"] = "VLLM (local models)"
+    except ImportError:
+        pass
+    
+    # Default fallback
+    if not providers:
+        providers["none"] = "No model providers available"
+    
+    return providers
+
+# ================================
+# HELPER FUNCTIONS
+# ================================
+
+def is_openai_available():
+    """Check if OpenAI is available"""
+    try:
+        api_key = settings.get('OPENAI_API_KEY', '')
+        return bool(api_key)
+    except:
+        return False
+
+def is_anthropic_available():
+    """Check if Anthropic is available"""
+    try:
+        api_key = settings.get('ANTHROPIC_API_KEY', '')
+        return bool(api_key)
+    except:
+        return False
+
+def is_openai_endpoint_available():
+    """Check if OpenAI endpoint is available"""
+    try:
+        api_key = settings.get('OPENAI_ENDPOINT_API_KEY', '')
+        return bool(api_key) 
+    except:
+        return False
+
+def is_ollama_available():
+    """Check if Ollama is running"""
+    try:
+        import requests
+        base_url = settings.get('OLLAMA_BASE_URL', settings.llm.get('ollama_base_url', 'http://localhost:11434'))
+        response = requests.get(f"{base_url}/api/tags", timeout=1.0)
+        return response.status_code == 200
+    except:
+        return False
+
+def is_vllm_available():
+    """Check if VLLM capability is available"""
+    try:
+        import torch
+        import transformers
+        return True
+    except ImportError:
+        return False
+
+def is_lmstudio_available():
+    """Check if LM Studio is available"""
+    try:
+        import lmstudio
+        import requests
+        lmstudio_url = settings.llm.get('lmstudio_url', 'http://localhost:1234')
+        # Try to connect to check if running
+        response = requests.get(f"{lmstudio_url}/health", timeout=1.0)
+        return response.status_code == 200
+    except:
+        return False
+
+def is_llamacpp_available():
+    """Check if LlamaCpp is available and configured"""
+    try:
+        from langchain_community.llms import LlamaCpp
+        model_path = settings.llm.get('llamacpp_model_path', '')
+        return bool(model_path) and os.path.exists(model_path)
+    except:
+        return False
+
+def get_available_providers():
+    """Get dictionary of available providers"""
+    return get_available_provider_types()
+
+# Log which providers are available
+AVAILABLE_PROVIDERS = get_available_providers()
+logger.info(f"Available providers: {list(AVAILABLE_PROVIDERS.keys())}")
+
+# Check if selected provider is available
+selected_provider = settings.llm.provider.lower()
+if selected_provider not in AVAILABLE_PROVIDERS and selected_provider != "none":
+    logger.warning(f"Selected provider {selected_provider} is not available.")

{local_deep_research-0.1.18 → local_deep_research-0.1.19}/src/local_deep_research/defaults/local_collections.toml

@@ -1,6 +1,3 @@
-# Default local document collections configuration
-# Each collection functions as an independent search engine
-
 # Project Documents Collection
 [project_docs]
 name = "Project Documents"
@@ -15,6 +12,9 @@ max_filtered_results = 5
 chunk_size = 1000
 chunk_overlap = 200
 cache_dir = "__CACHE_DIR__/local_search/project_docs"
+strengths = ["project documentation", "specifications", "internal documents"]
+weaknesses = ["no external information", "limited to organizational knowledge"]
+reliability = 0.9
 
 # Research Papers Collection
 [research_papers]
@@ -30,6 +30,9 @@ max_filtered_results = 5
 chunk_size = 800
 chunk_overlap = 150
 cache_dir = "__CACHE_DIR__/local_search/research_papers"
+strengths = ["academic research", "scientific papers", "scholarly content"]
+weaknesses = ["potentially outdated", "limited to collected papers"]
+reliability = 0.85
 
 # Personal Notes Collection
 [personal_notes]
@@ -44,4 +47,7 @@ max_results = 30
 max_filtered_results = 10
 chunk_size = 500
 chunk_overlap = 100
-cache_dir = "__CACHE_DIR__/local_search/personal_notes"
+cache_dir = "__CACHE_DIR__/local_search/personal_notes"
+strengths = ["personal knowledge", "notes", "private documents"]
+weaknesses = ["subjective content", "informal information"]
+reliability = 0.75

{local_deep_research-0.1.18 → local_deep_research-0.1.19}/src/local_deep_research/defaults/main.toml

@@ -1,23 +1,41 @@
 
 # Main configuration for Local Deep Research
 
+[web]
+port = 5000
+host = "0.0.0.0"
+debug = true
+
+[llm]
+# LLM provider (one of: ollama, openai, anthropic, vllm, openai_endpoint, lmstudio, llamacpp)
+provider = "ollama"
+# Model name
+model = "gemma3:12b"
+# Temperature
+temperature = 0.7
+# Maximum tokens
+max_tokens = 30000
+# OpenAI-compatible endpoint URL
+openai_endpoint_url = "https://openrouter.ai/api/v1"
+# LM Studio URL (default: http://localhost:1234)
+lmstudio_url = "http://localhost:1234"
+# LlamaCpp model path
+llamacpp_model_path = ""
+# LlamaCpp parameters
+llamacpp_n_gpu_layers = 1
+llamacpp_n_batch = 512
+llamacpp_f16_kv = true
+
 [general]
 # Directory for research outputs (relative to user data directory)
 output_dir = "research_outputs"
-
 # Knowledge accumulation approach (NONE, QUESTION, or ITERATION)
 knowledge_accumulation = "ITERATION"
-
 # Maximum context size for knowledge accumulation
 knowledge_accumulation_context_limit = 2000000
-
 # Enable fact checking (experimental, works better with large LLMs)
 enable_fact_checking = false
 
-[web]
-port = 5000
-host = "0.0.0.0"
-debug = true
 
 [search]
 # Search tool to use (auto, wikipedia, arxiv, duckduckgo, serp, google_pse, etc.)

local_deep_research-0.1.19/src/local_deep_research/web/static/sounds/README.md

@@ -0,0 +1,29 @@
+# Sound Files for Notifications
+
+This directory contains sound files used for notifications in the Deep Research application.
+
+## Required Files
+
+1. `success.mp3` - Played when research completes successfully
+2. `error.mp3` - Played when research fails or encounters an error
+
+## Sound Sources
+
+You can download copyright-free sound files from these sources:
+
+- [Freesound](https://freesound.org/)
+- [Free Sound Library](https://www.freesoundslibrary.com/)
+
+## Recommended Sounds
+
+### Success Sound
+- [Success Sound by grunz](https://freesound.org/people/grunz/sounds/109662/)
+- Direct download: https://freesound.org/data/previews/109/109662_945474-lq.mp3
+
+### Error Sound
+- [Error Sound by Autistic Lucario](https://freesound.org/people/Autistic%20Lucario/sounds/142608/)
+- Direct download: https://freesound.org/data/previews/142/142608_1840739-lq.mp3
+
+## Usage
+
+The application will automatically use these sounds when research tasks complete or fail, but only when the browser tab is not in focus. 

{local_deep_research-0.1.18 → local_deep_research-0.1.19}/src/local_deep_research/web_search_engines/engines/meta_search_engine.py

@@ -91,14 +91,23 @@ class MetaSearchEngine(BaseSearchEngine):
         if not self.available_engines:
             logger.warning("No search engines available")
             return []
-        
-        # Create engine descriptions for the prompt
-        engine_descriptions = "\n".join([
-            f"- {name.upper()}: Good for {', '.join(SEARCH_ENGINES[name]['strengths'][:3])}. "
-            f"Weaknesses: {', '.join(SEARCH_ENGINES[name]['weaknesses'][:2])}. "
-            f"Reliability: {SEARCH_ENGINES[name]['reliability']*100:.0f}%"
-            for name in self.available_engines
-        ])
+        engine_descriptions = []
+        for name in self.available_engines:
+            logger.info(f"Processing search engine: {name}")
+            try:
+                description = f"- {name.upper()}: Good for {', '.join(SEARCH_ENGINES[name]['strengths'][:3])}. " \
+                            f"Weaknesses: {', '.join(SEARCH_ENGINES[name]['weaknesses'][:2])}. " \
+                            f"Reliability: {SEARCH_ENGINES[name]['reliability']*100:.0f}%"
+                engine_descriptions.append(description)
+            except KeyError as e:
+                logger.error(f"Missing key for engine {name}: {e}")
+                # Add a basic description for engines with missing configuration
+                engine_descriptions.append(f"- {name.upper()}: General purpose search engine.")
+            except Exception as e:
+                logger.error(f"Error processing engine {name}: {e}")
+                engine_descriptions.append(f"- {name.upper()}: General purpose search engine.")
+
+        engine_descriptions = "\n".join(engine_descriptions)
         
         prompt = f"""Analyze this search query and rank the available search engines in order of most to least appropriate for answering it.